The determination of the California Bearing Ratio value of soil is tiresome, uneconomical, and time-consuming in the laboratory. Therefore, there is a required automation system to determine the California Bearing Ratio value of soil. Machine learning algorithms are being used for automation systems. In this paper, Artificial Neural Network has been proposed for the prediction of the California Bearing Ratio value of soil. Ash percentage, Liquid Limit, Plastic_Limit, Plasticity index, Shrinkage Limit, MDD and OMC parameters of soil affect the value of the California Bearing Ratio. In the laboratory, the training dataset has generated using these parameters of soil. The proposed classifier has been trained and tested using the training and testing dataset. Experimental results show that the proposed Artificial Neural Network is very accurate to predict California Bearing Ratio values of soil. It is also observed that the linear regression algorithm is very easy and useful to determine the value of the California Bearing ratio depending on seven attributes of soil. The rules generated by J48 and PART can be used to determine the California Bearing ratio. These models are very useful for civil engineers and civil constructors as a California Bearing ratio prediction automation system.